Log Level Reference

TRACE through FATAL - when to use each one, an example, and the mistake that trains on-call to ignore your alerts.

Something unexpected happened but the system recovered or degraded gracefully.

Retrying failed payment webhook (attempt 2/3)

Always on - a rising WARN rate is often the earliest sign of a real problem.

Was this tool helpful?

See this same kind of analysis applied to your whole repository

Call graphs, dead code, blast radius, and a health score. Results in 60 seconds, free.

Related tools

Alert Threshold Calculator

Baseline mean, std deviation, and a sigma multiplier to a statistically grounded alert threshold.

PromQL Rate Query Builder

Metric, labels, range window, aggregation - a syntactically valid rate/irate/increase query.

SLO Error Budget Calculator

SLO target, window, and downtime so far - your actual spendable error budget in minutes.

Prometheus Alert Rule Generator

Expression, threshold, and for-duration - a valid Prometheus alerting rule YAML.

Log Query Syntax Translator

The same filters, translated into Kibana KQL, Datadog, and Splunk SPL syntax side by side.

Runs entirely in your browser. Nothing is stored. Free forever.